Source code for olympus.datasets.split.constrained_bootstrap

from olympus.datasets.split.balanced_classes import balanced_random_indices, Split


[docs]def constrained_bootstrap_random_indices(rng, indices, n_train, n_valid, n_test):
    indices = set(indices)

    unique_train_indices = rng.choice(list(indices), replace=False, size=n_train)
    train_indices = rng.choice(unique_train_indices, replace=True, size=n_train)

    indices -= set(train_indices)

    valid_indices = rng.choice(list(indices), size=n_valid, replace=True)

    indices -= set(valid_indices)

    test_indices = rng.choice(list(indices), size=n_test, replace=True)

    indices -= set(test_indices)

    return Split(train=train_indices, valid=valid_indices, test=test_indices)


[docs]def split(datasets, data_size, seed, ratio, index):
    return balanced_random_indices(
        method=constrained_bootstrap_random_indices,
        classes=datasets.classes,
        n_points=data_size,
        seed=seed)